Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

Much of the literature in health economics has focused on examining patient and physician behavior within different institutional frameworks ultimately to suggest policies that will encourage the efficient provision of care. As patients, physicians and insurance providers are likely to have private information and competing objectives, deriving optimal insurance plans and physician payment schemes is complicated. Although much has been written on the topic, most of the results have been derived from a static model where illness is assumed to be a one-time event. However, these models may be limiting as patient health and physician-patient relationships are likely to evolve over time. For example, actions taken by General Practitioners (GPs) are likely to affect their patients' need for future care. That is, they are likely to affect patients' future consumption of medical services in general, but also their need for specialty or hospital care (which may be more complicated, invasive and costly). In response to this limitation, we propose a theoretical model where dynamics in the health production function are introduced in a meaningful way.

As noted above, the previous health-economics literature, which has focused on payment mechanisms, has mostly analyzed patient and physician behavior in a static framework. One can think of many scenarios, however, where the physician's current action may affect the patient's need for future care. Physician actions may include diagnostics, treatment and referrals. More specifically, the GP diagnostic ability allows him to identify mild cases from more serious ones. If a mild case is diagnosed, the GP may be able to treat the patient with preventative therapies which may in turn decrease the patient's likelihood of developing a more severe form of the illness and requiring more complex and costly care (such as specialty and in-hospital care). If a more severe case is diagnosed, the GP may direct the patient to specialty care in order to prevent future deteriorations in health, thus maximizing patient health and welfare, while potentially minimizing total health care expenditures.

In order to capture some of the features described above, we build a model where health can evolve over time and where GPs differ in their diagnosis ability and level of altruism (both of which are private information). GPs act as gatekeepers and must decide whether to treat their patient or refer them to specialty care. We solve for the optimal treatment and referral decisions for each physician as a function of their diagnostic ability, their level of altruism, as well as the likelihood that a low-severity illness left untreated worsens over time (what we call the First-Best outcome). We then consider several common types of physician payment mechanisms (including fee-for-service, capitation and fundholidng) and derive each physician's treatment and referral decision as a function of the same parameters. We then compare the treatment and referral decisions under the several payment mechanisms to the First-Best case. We show under which parameterization (i.e., for which diagnostic ability and altruism level) each of the payment mechanisms coincide with the First-Best.